- Email: [email protected]
The Impact of Radiation Therapy on the Risk of Lymphedema After Treatment for Breast Cancer: A Prospective Cohort Study
International Journal of
Radiation Oncology biology
physics
www.redjournal.org
Clinical Investigation
The Impact of Radiation Therapy on the Risk of Lymphedema After Treatment for Breast Cancer: A Prospective Cohort Study Laura E.G. Warren, MD,* Cynthia L. Miller, BS,* Nora Horick, MS,y Melissa N. Skolny, MSHA,* Lauren S. Jammallo, BS,* Betro T. Sadek, MD,* Mina N. Shenouda, MD,* Jean A. O’Toole, PT, MPH,z Shannon M. MacDonald, MD,* Michelle C. Specht, MD,x and Alphonse G. Taghian, MD, PhD* *Department of Radiation Oncology, yDepartment of Biostatistics, zDepartment of Physical and Occupational Therapy, and xDivision of Surgical Oncology, Massachusetts General Hospital, Boston, Massachusetts Received Oct 22, 2013, and in revised form Nov 10, 2013. Accepted for publication Nov 18, 2013.
Summary This prospective study evaluated the risk of lymphedema in women diagnosed and treated for breast cancer. It demonstrated that regional lymph node irradiation, with either inclusion of a supraclavicular field or a posterior axillary boost, increased the risk of lymphedema. These conclusions were based on a large cohort of patients treated with modern surgical and radiation therapy techniques and Perometer arm volume measurements. Close follow-up of women at risk for lymphedema will allow earlier detection and intervention.
Purpose/Objective: Lymphedema after breast cancer treatment can be an irreversible condition with a negative impact on quality of life. The goal of this study was to identify radiation therapy-related risk factors for lymphedema. Methods and Materials: From 2005 to 2012, we prospectively performed arm volume measurements on 1476 breast cancer patients at our institution using a Perometer. Treating each breast individually, 1099 of 1501 patients (73%) received radiation therapy. Arm measurements were performed preoperatively and postoperatively. Lymphedema was defined as 10% arm volume increase occurring >3 months postoperatively. Univariate and multivariate Cox proportional hazard models were used to evaluate risk factors for lymphedema. Results: At a median follow-up time of 25.4 months (range, 3.4-82.6 months), the 2-year cumulative incidence of lymphedema was 6.8%. Cumulative incidence by radiation therapy type was as follows: 3.0% no radiation therapy, 3.1% breast or chest wall alone, 21.9% supraclavicular (SC), and 21.1% SC and posterior axillary boost (PAB). On multivariate analysis, the hazard ratio for regional lymph node radiation (RLNR) (SC PAB) was 1.7 (PZ.025) compared with breast/chest wall radiation alone. There was no difference in lymphedema risk between SC and SC þ PAB (PZ.96). Other independent risk factors included early postoperative swelling (P<.0001), higher body mass index (P<.0001), greater number of lymph nodes dissected (PZ.018), and axillary lymph node dissection (PZ.0001). Conclusions: In a large cohort of breast cancer patients prospectively screened for lymphedema, RLNR significantly increased the risk of lymphedema compared with breast/chest wall radiation alone. When considering use of RLNR, clinicians should weigh the potential benefit of RLNR for control of disease against the increased risk of lymphedema. Ó 2014 Elsevier Inc.
Reprint requests to: Alphonse G. Taghian, MD, PhD, 100 Blossom Street, Boston, MA 02114. Tel: (617) 726-6050; E-mail: ataghian@ partners.org Presented at the 55th Annual Meeting of the American Society for Radiation Oncology (ASTRO), Atlanta, GA, September 22-25, 2013. Supported by Award # RO1CA139118 (A. Taghian) and Award # P50CA0893932 (A. Taghian) from the National Cancer Institute. The Int J Radiation Oncol Biol Phys, Vol. -, No. -, pp. 1e7, 2014 0360-3016/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijrobp.2013.11.232
content is solely the responsibility of the authors and does not necessarily represent the views of the National Cancer Institute or the National Institutes of Health. Conflict of interest: none. AcknowledgmentdThe authors thank Nadine R. Taghian, whose continuous dedication helped expedite completion of this project, and whose efforts are greatly appreciated.
2
Warren et al.
Introduction Women treated for breast cancer have increasingly successful outcomes; therefore, late toxicities of treatment are assuming greater importance. Lymphedema causes significant physical and psychosocial side effects and is of concern for women undergoing breast cancer therapy (1-4). Data regarding the risk of lymphedema with the inclusion of adjuvant radiation therapy are limited and are often based on retrospective cohorts, varying definitions of lymphedema, and heterogeneous patient populations. Of particular interest is the risk of lymphedema with extended radiation therapy fields. Regional lymph node radiation (RLNR) is routinely used in patients with 4 positive lymph nodes (LNs) found at the time of lumpectomy or mastectomy with axillary LN dissection (ALND) (5). These clinical practices are based on trials that demonstrated decreased rates of locoregional failure, increased disease-free survival (DFS), and increased overall survival (OS) for high-risk or LN-positive patients with the receipt of postmastectomy radiation therapy (6, 7). Inasmch as women undergoing RLNR most often also undergo sentinel LN biopsy (SLNB) or ALND, which have a reported incidence of lymphedema up to 11% and 30%, respectively (8-10), quantifying the additional risk with inclusion of RLNR has significant clinical implications. In this study, we used a large cohort of breast cancer patients prospectively screened for lymphedema and analyzed arm volume changes to quantify the risk of lymphedema with inclusion of adjuvant radiation therapy. The types of radiation therapy analyzed included receipt of no radiation, partial breast irradiation (PBI), whole breast irradiation (WBI) or chest wall radiation alone, or WBI/chest wall radiation with RLNR. Among patients who received RLNR, we sought to determine whether the addition of a posterior axillary boost (PAB) to supraclavicular irradiation (SC) increased the risk of lymphedema compared with SC alone.
Methods and Materials Patients We included 1476 women who underwent surgery for unilateral or bilateral breast cancer from 2005 to 2012. Patients were recruited at the time of initial consultation in a multidisciplinary clinic, and all underwent surgery at our institution. All patients, regardless of T or N stage, were included except for patients with pre-existing lymphedema. Treating each breast individually, we had 1501 unique cases. Each patient had >3 months of postsurgical follow-up.
Lymphedema definition and measurement According to the standard of care at our institution, all newly diagnosed breast cancer patients undergo screening for lymphedema with the use of serial Perometer arm volume measurements. A Perometer is an optoelectronic system that uses infrared beams to measure the limb and calculates the volume based on these measurements. Arm volume is measured preoperatively and at regular intervals corresponding with routine oncology follow-up visits during and after the completion of treatment. This protocol was approved by the Partners HealthCare Institutional Review Board and has been previously published (11).
International Journal of Radiation Oncology Biology Physics Lymphedema was defined as a 10% arm volume increase occurring >3 months postoperatively based on consensus in the literature (12, 13). For patients who underwent unilateral breast surgery, arm volume change was quantified by the relative volume change (RVC) equation (11). Briefly, RVC Z [(A(2)U(1)/U(2) A(1)) 1] where A(1), A(2) are preoperative (1) and postoperative (2) arm volumes on the surgical side and U(1), U(2) are arm volumes on the contralateral side at corresponding time points. This equation accounts for preoperative asymmetry between arms and uses contralateral arm volume as a control to account for factors unrelated to lymphedema, such as weight gain. The RVC equation is not applicable in patients who undergo bilateral breast surgery because a contralateral control arm is lacking. Therefore, the weight-adjusted volume change (WAC) equation was used for patients who underwent bilateral breast surgery (14). Briefly, WAC Z [(A(2)W(1)/W(2)A(1)) 1] where A(1), A(2) are preoperative (1) and postoperative (2) arm volumes on the surgical side, and W(1), W(2) are the patient’s weight at the corresponding time points. The equivalency of 10% according to the RVC and WAC equations as a criterion for lymphedema has been previously demonstrated (14). Arm volume change was calculated at each follow-up visit. Measurements obtained 3 months after surgery were not used to determine lymphedema incidence. A 10% arm volume increase occurring during this period was considered early postoperative swelling and was included as a separate variable in the analysis.
Radiation treatment Of the 1099 breasts that received radiation, 796 (72%) were at the Massachusetts General Hospital (MGH), 92 (8%) at MGHaffiliated institutions, and 211 (19%) at outside institutions. Each breast was categorized as follows: no radiation; PBI, WBI, or chest wall radiation alone (B/CW); or B/CW with RLNR. RLNR was classified as SC only or SC with a PAB. After 2008, supraclavicular level I, II, and III axillary LNs at our institution were contoured most of the time according to the Radiation Therapy Oncology Group atlas (15). The lateral border of the SC field was defined by the treating physician. The SC and level III LNs received a dose of 50 to 50.4 Gy in 25 to 28 fractions (1.82.0 Gy per fraction) through an anterior oblique field angled 10 to 15 to avoid the spinal cord. The dose was calculated at a depth of mostly 3 cm or a depth to include the majority of the contoured supraclavicular area, which varied from 2 to 5 cm. The PAB was completed by use of a posterior approach, and the dose was calculated at midplane. Photon energies of 6- or 10-MV photons were typically used. Until 2009, patients who received PBI were enrolled in a dose-escalation clinical trial (16) and received a total dose of 32, 36, or 40 Gy in 4 Gy per fraction twice daily with a 3-dimensional conformal technique that used a combination of photons and electrons (17). Thereafter, patients were treated off trial with 36 Gy in 9 fractions twice daily. The radiation specifics are given in Table 1.
Statistical analysis Univariate and multivariate analyses were used to identify independent predictors of lymphedema by use of the Cox proportional hazards method. The factors evaluated in these analyses included patient-related factors (age at diagnosis, body mass index [BMI],
Volume - Number - 2014 Table 1
Breast cancer irradiation and lymphedema
Radiation dose and fractionation by treatment field Radiation
Median dose, Gy (range)
Fractions (range)
Partial breast 36.0 (32.0-40.0) 9 Whole breast/chest wall tangents 50.0 (41.9-61.2) 25 Whole breast/chest wall boost 10.0 (2.5-20.0) 5 Supraclavicular 50.0 (42.0-50.4) 25 Posterior axillary boost 46.2 (35.0-48.6) 28
(8-10) (16-28) (3-10) (22-28) (22-28)
and presence of postoperative swelling, defined as a 10% arm volume increase 3 months from surgery); tumor-related factors (tumor size and number of pathologically involved LNs); and treatment-related factors (type of breast surgery, management of the axilla, number of LNs removed, use of systemic therapy, timing of systemic therapy, hormonal therapy, and radiation field [s]). The factors included were based on a review of prior literature regarding lymphedema-associated risk factors. Age at diagnosis, BMI, tumor size, the number of LNs dissected, and the number of pathologically involved LNs were analyzed as continuous variables in such a manner that the hazard ratios computed for these variables reflected the change in lymphedema risk associated with a 1-unit increase in the variable. Timedependent covariates were included for use of systemic therapies and radiation fields in such a manner that cases were included in the unexposed group before the initiation of a given treatment and in the exposed group thereafter. All the aforementioned factors were included in the initial univariate model. Only variables that were significant in the univariate analysis were included in the multivariate analysis. Tests of interaction were performed between the radiation groups and the number of nodes removed, number of positive nodes, and breast surgery, and there were no significant interactions. The Kaplan-Meier method was used to estimate and plot the cumulative incidence of lymphedema by radiation field. Patients were censored at the time of last follow-up measurement.
Results Of the 1476 patients, 1262 (86%) underwent unilateral breast surgery and 214 (14%) underwent bilateral breast surgery for unilateral (nZ189) or bilateral (nZ25) breast cancer. Postoperative radiation therapy was given after 1099 of 1501 (73%) individual breast surgeries. The clinical and pathologic characteristics are given in Table 2. The median postoperative follow-up time was 25.4 months (range, 3.0-82.6 months), with a median of 4 postoperative measurements per patient (range, 1-25 measurements).
Table 2 cohort
3
Clinical and pathologic characteristics of the study
Characteristic
Median (range)
nZ1501 (100%)
Clinical characteristics Age at diagnosis, y 54 (23-89) BMI at diagnosis, kg/m2 26.1 (16.5-59.0) Pathologic characteristics Invasive tumor size,* cm 1.5 (0.01-10.5) Breast surgery Lumpectomy 927 (62) Mastectomy 574 (38) Axillary surgery None 178 (12) SLNB 928 (62) ALND 395 (26) No. of LNs removed, SLNB 2 (1-13) No. of LNs removed, ALND 16 (3-43) No. of positive LNs, ALND 2 (0-39) Radiation therapy None 402 (27) Partial breast 92 (6) B/CW 698 (47) B/CW þ SC 194 (13) B/CW þ SC þ PAB 115 (8) Neoadjuvant chemotherapy Yes 151 (10) No 1350 (90) Adjuvant chemotherapy Yes 571 (38) No 930 (62) Hormonal therapy Yes 1110 (74) No 391 (26) Abbreviations: ALND Z axillary lymph node dissection; B/CW Z whole breast/chest wall only; BMI Z body mass index; LN Z lymph node; PAB Z posterior axillary boost; SC Z supraclavicular; SLNB Z sentinel lymph node biopsy. * Excludes tumor size for patients who underwent neoadjuvant chemotherapy.
The 2-year cumulative incidence of lymphedema by radiation field and axillary surgery is shown in Table 3. For SLNB, the 2year cumulative incidence was 6.1% (95% CI, 1.5%-23.4%) with RLNR (nZ46) and 2.8% (95% CI, 1.6%-4.8%) with PBI/ WBI/chest wall alone (nZ569). For ALND, the incidence was 24.3% (95% CI, 18.9%-30.9%) with RLNR (nZ262) and 7.3% (95% CI, 3.1%-16.8%) with WBI/chest wall alone (nZ73).
Lymphedema incidence Univariate and multivariate analysis The overall cumulative incidence of lymphedema increased from 6.8% (95% confidence interval [CI], 5.5%-8.4%) at 24 months to 13.7% (95% CI, 11.1%-16.9%) at 60 months postoperatively (Fig. 1). The median time to the development of lymphedema was 24.1 months postoperatively (range, 3.0-82.6 months). The 2-year cumulative incidence of lymphedema was higher with RLNR as SC (21.9%; 95% CI, 16.0%-29.5%) or SC þ PAB (21.1%; 95% CI, 13.8%-31.5%) than with no radiation (3.0%; 95% CI, 1.6%-5.6%) or PBI/WBI/chest wall radiation alone (3.1%; 95% CI, 2.0%-4.7%).
The results of univariate analysis are included in Table 4. In a multivariate model using no radiation as the reference group, PBI (PZ.67) and WBI/chest wall radiation alone (PZ.92) did not significantly increase lymphedema risk. Given these results, a multivariate model was constructed with PBI and WBI/chest wall radiation groups combined as the reference group based on perceived clinical relevance. Additionally, because the type of breast surgery was not significant in the multivariate model (PZ.99) and there was not a significant interaction between type
4
International Journal of Radiation Oncology Biology Physics
Warren et al.
Fig. 1. Cumulative incidence of lymphedema stratified by type of radiation therapy. B/CW Z breast/chest wall only; PAB Z posterior axillary boost; SC Z supraclavicular. of breast surgery and radiation, all patients, regardless of surgical type, were included in the final multivariate model. The results of the multivariate analysis are seen in Table 5. The addition of RLNR (SC or SC þ PAB) when compared with breast/chest wall radiation alone significantly increased lymphedema risk, with a hazard ratio (HR) of 1.70 (95% CI, 1.07-2.70, PZ.025). Other factors significant for increased lymphedema risk on multivariate analysis included early postoperative swelling (HR Z 10.3, P<.0001), higher preoperative BMI (HR Z 1.06, P<.0001), number of axillary LNs dissected (HR Z 1.04, PZ.02), and undergoing ALND (HR Z 3.5, PZ.0001).
Discussion The current study used a cohort of 1476 breast cancer patients prospectively screened for lymphedema at a single institution. The patients in our study were treated for breast cancer between 2005 and 2012, and they thereby represent a population treated with modern radiation therapy and surgical techniques over a relatively short time. The risk factors for lymphedema, including adjuvant radiation therapy, have been previously evaluated; however, most studies have used retrospectively collected data, variable and Table 3
largely subjective methods, or both to measure and define lymphedema (12, 18-20). The cumulative incidence of lymphedema in our series was 6.8% at 24 months and 13.7% at 60 months postoperatively. This is lower than has been reported in previous studies, likely because we included a large cohort of women who did not undergo axillary surgery (nZ178). When stratified by radiation therapy field, the 2-year cumulative incidence was 3.0%, 3.1%, 21.9%, and 21.1% for no radiation, breast/chest wall radiation alone, B/CW þ SC, and B/CW þ SC þ PAB, respectively. To evaluate the risk of lymphedema with inclusion of RLNR, we used a reference group composed of patients who received PBI or WBI/chest wall radiation alone, given that RLNR is unlikely to be used without also irradiating the breast or chest wall. In our series, the inclusion of a SC field, with or without the inclusion of a PAB, significantly increased the risk of lymphedema compared with breast/chest wall radiation alone. The addition of a PAB to a SC field did not increase the risk of lymphedema in comparison with a SC field alone. When further stratified by type of axillary surgery, for patients who underwent ALND, the 2-year cumulative incidence was higher with RLNR (as SC or SC þ PAB) than with breast/chest wall radiation alone, 24.3% versus 7.3%, respectively. A similar trend was observed for patients who underwent SLNB (2.8% without RLNR vs 6.1% with RLNR); however, confidence intervals were overlapping because of a small number of patients who underwent SLNB with RLNR (nZ46). These findings are consistent with several prior studies. In a retrospective study of 727 patients treated from 1982 to 1995, Coen et al (19) used a postoperative arm circumference measurement (without preoperative assessment) to evaluate predictors of lymphedema in patients undergoing breast conservation therapy (BCT). The authors found that the addition of RLNR significantly increased the risk of lymphedema from 1.8% to 8.9% with inclusion of RLNR (SC with or without PAB). Another study by Hayes et al (18) evaluated a cohort of 2579 breast cancer patients treated between 1970 and 2005, classifying lymphedema based on physician’s examination, without preoperative assessment. Consistently with our findings, the authors concluded that the inclusion of RLNR significantly increased the risk of lymphedema (PZ.0001), with no significant difference between SC and SC þ PAB (PZ.1). By contrast, Shah et al (20) studied 1497 patients treated from 1980 to 2006 with BCT and did not find that RLNR was a significant risk factor for lymphedema. Increased rates with SC (9.9% vs 7.0%, PZ.29) and PAB (14.7% vs 7.3%, PZ.40) were not observed compared with breast radiation alone. The study did not use a preoperative assessment, and it defined lymphedema based on physical examination with or without circumference measurements. Graham et al (21) evaluated the impact of SC radiation therapy volumes on lymphedema in a cross-sectional study
Cumulative incidence of lymphedema 24 months postoperatively stratified by type of radiation therapy and axillary surgery Cumulative incidence of lymphedema*
Radiation Radiation field(s) No radiation B/CW only B/CW þ SC B/CW þ SC þ PAB
Entire cohort (nZ1501) 3.0% (1.6-5.6) (nZ402) 3.1% (2.0-4.7) (nZ790) 21.9% (16.0-29.5) (nZ194) 21.1% (13.8-31.5) (nZ115)
SLNB 0.00% 2.82% 7.14% 4.76%
(nZ928) (0.00-0.00) (nZ313) (1.7-4.8) (nZ569) (1.0-40.9) (nZ25) (0.7-29.3) (nZ21)
ALND (nZ395) 18.27% (9.7-32.9) (nZ60) 7.33% (3.1-16.8) (nZ73) 23.93% (17.5-32.2) (nZ169) 24.89% (16.2-37.1) (nZ94)
Abbreviations: ALND Z axillary lymph node dissection; B/CW Z partial breast or whole breast/chest wall; PAB Z posterior axillary boost; SC Z supraclavicular; SLNB Z sentinel lymph node biopsy. * 95% confidence interval appears after percentage. No. in each subgroup is shown.
Volume - Number - 2014
Breast cancer irradiation and lymphedema
Table 4 Univariate results for association of factors with risk of lymphedema Variable Clinical characteristics Age at diagnosis* BMI* Pathologic characteristics Invasive tumor size* Breast surgery Mastectomy vs lumpectomy Axillary surgery SLNB vs no axillary surgery ALND vs SLNB No. of LNs removed* No. of positive LNs* Radiation therapyy Partial breast only Whole breast/chest wall only Whole breast/chest wall þ SC Whole breast/chest wall þ SC þ PAB Systemic therapy Neoadjuvant chemotherapy Adjuvant chemotherapy Hormonal therapy Early postoperative swelling 10%, 3 months postoperatively
Hazard 95% Confidence ratio interval
P value
1.01 1.06
0.99-1.02 1.04-1.09
.38 <.0001
1.32
1.22-1.44
<.0001
1.87
1.31-2.67
.0006
0.74
0.31-1.81
.51
9.17 1.10 1.09
5.88-14.29 1.08-1.11 1.05-1.13
<.0001 <.0001 <.0001
0.38 0.85
0.09-1.56 0.51-1.42
.18 .53
5.05
3.11-8.19
<.0001
4.98
2.92-8.47
<.0001
2.61
1.68-4.07
<.0001
2.35 1.68
1.62-3.39 1.13-2.51
<.0001 .01
6.06-26.1
<.0001
12.6
Abbreviations: ALND Z axillary lymph node dissection; BMI Z body mass index; LN Z lymph node; SC Z supraclavicular; SLNB Z sentinel lymph node biopsy; PAB Z posterior axillary boost. * Analyzed as continuous variables; hazard ratio reflects the change in lymphedema risk associated with a 1-unit increase. y Reference group: no radiation therapy.
of 91 breast cancer patients who had postoperative (but no preoperative) arm circumference measurements. They found that SC volumes limited laterally by the coracoid process did not increase the lymphedema risk expected from ALND, whereas expanding the field or including a PAB did. Our cohort uniquely included 402 patients who did not receive radiation, primarily mastectomy patients with limited nodal involvement and older women undergoing lumpectomy. When this cohort was used as a reference group, PBI and WBI/chest wall radiation alone did not significantly increase lymphedema risk. Nonreceipt of radiation therapy is rarely a comparison group in other studies; however, our findings are consistent with those from Liljegren et al (22), which demonstrated that postoperative breast radiation therapy alone did not contribute to lymphedema. It has been suggested that the use of PBI may reduce the rates of lymphedema secondary to lower doses to the axilla (23). Our series demonstrated low rates for patients who underwent BCT with PBI or WBI, whether after SLNB or no axillary surgery. The PBI cohort
5
Table 5 Multivariate results for factors associated with risk of lymphedema Variable No radiation* RLNR* B/CW þ SC B/CW þ SC þ PAB ALND No. of LNs removedy BMIy 10% swelling, 3 months postoperatively
Hazard 95% Confidence ratio interval 0.99 1.70 1.68 1.72 3.52 1.04 1.06 10.3
0.59-1.67 1.07-2.70 1.00-2.83 1.00-2.96 1.84-6.73 1.01-1.07 1.03-1.09 4.52-23.3
P value .9825 .0253 .0492 .0484 .0001 .0184 <.0001 <.0001
Abbreviations: ALND Z axillary lymph node dissection; BMI Z body mass index; LN Z lymph node; RLNR Z regional lymph node radiation. * Radiation reference group: Partial breast and whole breast/chest wall only. y Analyzed as continuous variables; hazard ratio reflects the change in lymphedema risk associated with a 1-unit increase.
in our study was composed of patients who underwent lumpectomy for low-risk, node-negative disease; therefore, further research is warranted to determine whether PBI reduces lymphedema risk compared with WBI for patients with higher-risk or node-positive disease, as in the NSABP B-39/Radiation Therapy Oncology Group 0413 trial (Clinicaltrials.gov ID #: NCT00103181). Other independent risk factors for lymphedema in our series included higher preoperative BMI, ALND, and early postoperative swelling. A previous study by Mahamaneerat et al (24) evaluated postoperative arm swelling as a risk factor for lymphedema and found that a 5% increase in arm volume 1 month after surgery was associated with a 1.4-fold increased risk of lymphedema. Interestingly, in this study, early postoperative swelling conferred the highest risk of lymphedema for all variables analyzed, with a hazard ratio of 10.3. Higher BMI and ALND have also been consistently reported as risk factors in previous studies (8, 12, 25, 26). We found that even after type of axillary surgery was controlled for in a multivariate model, lymphedema risk remained significantly increased with a greater number of LNs removed, which has also been demonstrated in prior studies (18, 25). Although these data are unlikely to change the SLNB or ALND procedure, it may allow for the identification of women for whom closer surveillance is appropriate. The findings from several recent studies may have an impact on practices regarding treatment of the axilla. The ACOSOG Z0011 trial demonstrated no difference in the rates of regional recurrence, DFS, or OS in 891 women with limited nodal involvement receiving SLNB randomized to completion ALND or no further axillary treatment (27, 28). Although it was recommended that patients not receive a third field, they may have had deep or partially deep tangents, which can cover a significant part of the axilla (29). The rates of lymphedema in this trial have not yet been reported. Preliminary findings presented by Rutgers et al (30) on the EORTC 10981-22023 AMAROS [After Mapping of the Axilla: Radiotherapy Or Surgery] phase 3 trial comparing axillary radiation with ALND after a positive SLNB result demonstrated no differences in OS or DFS. The lymphedema rates were higher for ALND (28%) without RLNR than for SLNB with RLNR (14%) at 5 years (P<.0001). Whelan et al (31) reported on the NCIC-CTG MA.20 Intergroup trial, which randomized high-risk women (defined by
6
Warren et al.
positive LN or negative LN with high-risk features) who underwent BCT to receive WBI with or without RLNR. The inclusion of RLNR improved locoregional DFS and distant DFS, and it trended toward a statistically significant improvement in OS. The rate of lymphedema was 4.1% and 7.3% for WBI and WBI þ RLNR, respectively (PZ.004). We similarly observed an increased lymphedema incidence for patients who underwent ALND and RLNR (24.3%) compared with ALND and B/CW radiation alone (7.3%) and/or SLNB and RLNR (6.1%), although these rates include mastectomy patients, unlike the MA.20 trial, and the number of patients in each subgroup is small. A better understanding of the morbidities secondary to combinations of surgical and radiation therapy treatments of the axilla is increasingly important because of the likely change in practice patterns resulting from the publication of these trials, including more frequent use of RLNR. Our study has potential limitations. Potential bias is associated with nonrandom assignment of radiation fields. Additionally, on the basis of practice patterns, the number of patients who underwent SLNB and received RLNR was small, making subgroup analysis challenging and warranting further investigation. Last, the size of the supraclavicular field, and in particular the lateral border, was not examined in the current study. This may have an impact on lymphedema risk and will be the focus of further research. Our study has several strengths. The patients underwent preoperative and regular postoperative measurements by use of a Perometer. The validity and accuracy of the Perometer in quantifying arm volume has been rigorously demonstrated (32). The importance of obtaining preoperative assessments to account for asymmetry between arms, and of incorporating change in contralateral arm size to account for factors unrelated to lymphedema, has also been demonstrated (11, 14, 33, 34). The RVC and WAC equations used in this study account for both of these factors. Previous often-cited studies evaluating radiation therapy-related risk factors for lymphedema have used patients who underwent breast cancer treatment over several decades, and they lack prospectively collected and accurate assessments with incorporation of a preoperative assessment (18-20). By contrast, the current study used patients treated over a 7-year period and reflects modern surgical and radiation therapy techniques, including an increasing use of SLNB as opposed to ALND, and inclusion of patients who received PBI and those with positive SLNB who received RLNR instead of ALND. In conclusion, the addition of RLNR, either with SC or SC þ PAB, significantly increased the risk of lymphedema when compared with breast or chest wall radiation alone. Other independent risk factors included higher preoperative BMI, use of ALND, number of LN dissected, and early postoperative swelling. When considering the inclusion of RLNR, clinicians should weigh the potential benefit of RLNR for control of disease against the increased risk of lymphedema, particularly in women with other potential risk factors for lymphedema. Women who do receive RLNR should be prospectively monitored for lymphedema to ensure early detection and possible intervention. Future studies should further evaluate the risk of lymphedema after SLNB with RLNR compared with completion ALND in the setting of BCT or mastectomy.
References 1. Ahmed RL, Schmitz KH, Prizment AE, et al. Risk factors for lymphedema in breast cancer survivors: The Iowa Women’s Health Study. Breast Cancer Res Treat 2011;130:981-991.
International Journal of Radiation Oncology Biology Physics 2. Ja¨ger G, Do¨ller W, Roth R. Quality-of-life and body image impairments in patients with lymphedema. Lymphology 2006;39:193-200. 3. Tobin MB, Lacey HJ, Meyer L, et al. The psychological morbidity of breast cancer-related arm swelling: Psychological morbidity of lymphoedema. Cancer 1993;72:3248-3252. 4. Khan F, Amatya B, Pallant JF, et al. Factors associated with long-term functional outcomes and psychological sequelae in women after breast cancer. Breast 2012;21:314-320. 5. Ceilley E, Jagsi R, Goldberg S, et al. Radiotherapy for invasive breast cancer in North America and Europe: Results of a survey. Int J Radiat Oncol Biol Phys 2005;61:365-373. 6. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997;337:949-955. 7. Ragaz J, Olivotto IA, Spinelli JJ, et al. Locoregional radiation therapy in patients with high-risk breast cancer receiving adjuvant chemotherapy: 20-year results of the British Columbia randomized trial. J Natl Cancer Inst 2005;97:116-126. 8. McLaughlin SA, Wright MJ, Morris KT, et al. Prevalence of lymphedema in women with breast cancer 5 years after sentinel lymph node biopsy or axillary dissection: Objective measurements. J Clin Oncol 2008;26:5213-5219. 9. Wilke LG, McCall LM, Posther KE, et al. Surgical complications associated with sentinel lymph node biopsy: Results from a prospective international cooperative group trial. Ann Surg Oncol 2006;13:491-500. 10. Langer I, Guller U, Berclaz G, et al. Morbidity of sentinel lymph node biopsy (SLN) alone versus SLN and completion axillary lymph node dissection after breast cancer surgery: A prospective Swiss multicenter study on 659 patients. Ann Surg 2007;245:452-461. 11. Ancukiewicz M, Russell TA, Otoole J, et al. Standardized method for quantification of developing lymphedema in patients treated for breast cancer. Int J Radiat Oncol Biol Phys 2011;79:1436-1443. 12. Tsai RJ, Dennis LK, Lynch CF, et al. The risk of developing arm lymphedema among breast cancer survivors: A meta-analysis of treatment factors. Ann Surg Oncol 2009;16:1959-1972. 13. Armer JM, Stewart BR. A comparison of four diagnostic criteria for lymphedema in a post-breast cancer population. Lymphat Res Biol 2005;3:208-217. 14. Miller CL, Specht MC, Horick N, et al. A novel, validated method to quantify breast-cancer related lymphedema (BCRL) following bilateral breast surgery. Lymphology 2013;46:64-74. 15. Li XA, Tai A, Arthur DW, et al. Variability of target and normal structure delineation for breast cancer radiotherapy: An RTOG multiinstitutional and multiobserver study. Int J Radiat Oncol Biol Phys 2009;73:944-951. 16. Pashtan IM, Recht A, Ancukiewicz M, et al. External beam accelerated partial-breast irradiation using 32 gy in 8 twice-daily fractions: 5year results of a prospective study. Int J Radiat Oncol Biol Phys 2012; 84:e271-e277. 17. Taghian AG, Kozak KR, Katz A, et al. Accelerated partial breast irradiation using proton beams: Initial dosimetric experience. Int J Radiat Oncol Biol Phys 2006;65:1404-1410. 18. Hayes SB, Freedman GM, Li T, et al. Does axillary boost increase lymphedema compared with supraclavicular radiation alone after breast conservation? Int J Radiat Oncol Biol Phys 2008;72:1449-1455. 19. Coen JJ, Taghian AG, Kachnic LA, et al. Risk of lymphedema after regional nodal irradiation with breast conservation therapy. Int J Radiat Oncol Biol Phys 2003;55:1209-1215. 20. Shah C, Wilkinson JB, Baschnagel A, et al. Factors associated with the development of breast cancer-related lymphedema after whole-breast irradiation. Int J Radiat Oncol Biol Phys 2012;83:1095-1100. 21. Graham P, Jagavkar R, Browne L, et al. Supraclavicular radiotherapy must be limited laterally by the coracoid to avoid significant adjuvant breast nodal radiotherapy lymphoedema risk. Australas Radiol 2006; 50:578-582. 22. Liljegren G, Holmberg L. Arm morbidity after sector resection and axillary dissection with or without postoperative radiotherapy in breast
Volume - Number - 2014
23. 24.
25.
26.
27.
28.
cancer stage I. Results from a randomised trial. Uppsala-Orebro Breast Cancer Study Group. Eur J Cancer 1997;33:193-199. Shah C, Badiyan S, Khwaja S, et al. Breast cancer related lymphedema: A review of recent developments. Androl Gynecol Curr Res 2013;1:2. Mahamaneerat WK, Shyu C-R, Stewart BR, et al. Breast cancer treatment, BMI, post-op swelling/lymphoedema. J Lymphoedema 2008;3:38-44. Disipio T, Rye S, Newman B, et al. Incidence of unilateral arm lymphoedema after breast cancer: A systematic review and metaanalysis. Lancet Oncol 2013;14:500-515. Lucci A, McCall LM, Beitsch PD, et al. Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J Clin Oncol 2007; 25:3657-3663. Giuliano AE, Hunt KK, Ballman KV, et al. Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: A randomized clinical trial. JAMA 2011;305:569-575. Giuliano AE, McCall L, Beitsch P, et al. Locoregional recurrence after sentinel lymph node dissection with or without axillary dissection in patients with sentinel lymph node metastases: The American College of Surgeons Oncology Group Z0011 randomized trial. Ann Surg 2010; 252:426-432. discussion 432-433.
Breast cancer irradiation and lymphedema
7
29. Haffty BG, Hunt KK, Harris JR, et al. Positive sentinel nodes without axillary dissection: Implications for the radiation oncologist. J Clin Oncol 2011;29:4479-4481. 30. Rutgers EJ, Donker M, Straver ME, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer patients: Final analysis of the EORTC AMAROS trial (10981/22023) [Abstract]. In: Proceedings of the ASCO Annual Meeting; 2013 May 31-June 4; Chicago, IL. J Clin Oncol 2013; 31 (Suppl.):Abstr LBA1001. 31. Whelan T, Olivotto I, Chapman J. NCIC-CTG MA.20: An intergroup trial of regional nodal irradiation in early breast cancer [Abstract]. In: Proceedings of the ASCO Annual Meeting; 2011 June 3-June 7; Chicago, IL. J Clin Oncol 2011;29 (Suppl.):Abstr LBA 1003. 32. Tierney S, Aslam M, Rennie K, et al. Infrared optoelectronic volumetry: The ideal way to measure limb volume. Eur J Vasc Endovasc Surg 1996;12:412-417. 33. Stout Gergich NL, Pfalzer LA, McGarvey C, et al. Preoperative assessment enables the early diagnosis and successful treatment of lymphedema. Cancer 2008;112:2809-2819. 34. Ancukiewicz M, Miller CL, Skolny MN, et al. Comparison of relative versus absolute arm size change as criteria for quantifying breast cancer-related lymphedema: The flaws in current studies and need for universal methodology. Breast Cancer Res Treat 2012;135:145-152. erratum in Breast Cancer Res Treat 2012;136:623.
Radiation Oncology biology
physics
www.redjournal.org
Clinical Investigation
The Impact of Radiation Therapy on the Risk of Lymphedema After Treatment for Breast Cancer: A Prospective Cohort Study Laura E.G. Warren, MD,* Cynthia L. Miller, BS,* Nora Horick, MS,y Melissa N. Skolny, MSHA,* Lauren S. Jammallo, BS,* Betro T. Sadek, MD,* Mina N. Shenouda, MD,* Jean A. O’Toole, PT, MPH,z Shannon M. MacDonald, MD,* Michelle C. Specht, MD,x and Alphonse G. Taghian, MD, PhD* *Department of Radiation Oncology, yDepartment of Biostatistics, zDepartment of Physical and Occupational Therapy, and xDivision of Surgical Oncology, Massachusetts General Hospital, Boston, Massachusetts Received Oct 22, 2013, and in revised form Nov 10, 2013. Accepted for publication Nov 18, 2013.
Summary This prospective study evaluated the risk of lymphedema in women diagnosed and treated for breast cancer. It demonstrated that regional lymph node irradiation, with either inclusion of a supraclavicular field or a posterior axillary boost, increased the risk of lymphedema. These conclusions were based on a large cohort of patients treated with modern surgical and radiation therapy techniques and Perometer arm volume measurements. Close follow-up of women at risk for lymphedema will allow earlier detection and intervention.
Purpose/Objective: Lymphedema after breast cancer treatment can be an irreversible condition with a negative impact on quality of life. The goal of this study was to identify radiation therapy-related risk factors for lymphedema. Methods and Materials: From 2005 to 2012, we prospectively performed arm volume measurements on 1476 breast cancer patients at our institution using a Perometer. Treating each breast individually, 1099 of 1501 patients (73%) received radiation therapy. Arm measurements were performed preoperatively and postoperatively. Lymphedema was defined as 10% arm volume increase occurring >3 months postoperatively. Univariate and multivariate Cox proportional hazard models were used to evaluate risk factors for lymphedema. Results: At a median follow-up time of 25.4 months (range, 3.4-82.6 months), the 2-year cumulative incidence of lymphedema was 6.8%. Cumulative incidence by radiation therapy type was as follows: 3.0% no radiation therapy, 3.1% breast or chest wall alone, 21.9% supraclavicular (SC), and 21.1% SC and posterior axillary boost (PAB). On multivariate analysis, the hazard ratio for regional lymph node radiation (RLNR) (SC PAB) was 1.7 (PZ.025) compared with breast/chest wall radiation alone. There was no difference in lymphedema risk between SC and SC þ PAB (PZ.96). Other independent risk factors included early postoperative swelling (P<.0001), higher body mass index (P<.0001), greater number of lymph nodes dissected (PZ.018), and axillary lymph node dissection (PZ.0001). Conclusions: In a large cohort of breast cancer patients prospectively screened for lymphedema, RLNR significantly increased the risk of lymphedema compared with breast/chest wall radiation alone. When considering use of RLNR, clinicians should weigh the potential benefit of RLNR for control of disease against the increased risk of lymphedema. Ó 2014 Elsevier Inc.
Reprint requests to: Alphonse G. Taghian, MD, PhD, 100 Blossom Street, Boston, MA 02114. Tel: (617) 726-6050; E-mail: ataghian@ partners.org Presented at the 55th Annual Meeting of the American Society for Radiation Oncology (ASTRO), Atlanta, GA, September 22-25, 2013. Supported by Award # RO1CA139118 (A. Taghian) and Award # P50CA0893932 (A. Taghian) from the National Cancer Institute. The Int J Radiation Oncol Biol Phys, Vol. -, No. -, pp. 1e7, 2014 0360-3016/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijrobp.2013.11.232
content is solely the responsibility of the authors and does not necessarily represent the views of the National Cancer Institute or the National Institutes of Health. Conflict of interest: none. AcknowledgmentdThe authors thank Nadine R. Taghian, whose continuous dedication helped expedite completion of this project, and whose efforts are greatly appreciated.
2
Warren et al.
Introduction Women treated for breast cancer have increasingly successful outcomes; therefore, late toxicities of treatment are assuming greater importance. Lymphedema causes significant physical and psychosocial side effects and is of concern for women undergoing breast cancer therapy (1-4). Data regarding the risk of lymphedema with the inclusion of adjuvant radiation therapy are limited and are often based on retrospective cohorts, varying definitions of lymphedema, and heterogeneous patient populations. Of particular interest is the risk of lymphedema with extended radiation therapy fields. Regional lymph node radiation (RLNR) is routinely used in patients with 4 positive lymph nodes (LNs) found at the time of lumpectomy or mastectomy with axillary LN dissection (ALND) (5). These clinical practices are based on trials that demonstrated decreased rates of locoregional failure, increased disease-free survival (DFS), and increased overall survival (OS) for high-risk or LN-positive patients with the receipt of postmastectomy radiation therapy (6, 7). Inasmch as women undergoing RLNR most often also undergo sentinel LN biopsy (SLNB) or ALND, which have a reported incidence of lymphedema up to 11% and 30%, respectively (8-10), quantifying the additional risk with inclusion of RLNR has significant clinical implications. In this study, we used a large cohort of breast cancer patients prospectively screened for lymphedema and analyzed arm volume changes to quantify the risk of lymphedema with inclusion of adjuvant radiation therapy. The types of radiation therapy analyzed included receipt of no radiation, partial breast irradiation (PBI), whole breast irradiation (WBI) or chest wall radiation alone, or WBI/chest wall radiation with RLNR. Among patients who received RLNR, we sought to determine whether the addition of a posterior axillary boost (PAB) to supraclavicular irradiation (SC) increased the risk of lymphedema compared with SC alone.
Methods and Materials Patients We included 1476 women who underwent surgery for unilateral or bilateral breast cancer from 2005 to 2012. Patients were recruited at the time of initial consultation in a multidisciplinary clinic, and all underwent surgery at our institution. All patients, regardless of T or N stage, were included except for patients with pre-existing lymphedema. Treating each breast individually, we had 1501 unique cases. Each patient had >3 months of postsurgical follow-up.
Lymphedema definition and measurement According to the standard of care at our institution, all newly diagnosed breast cancer patients undergo screening for lymphedema with the use of serial Perometer arm volume measurements. A Perometer is an optoelectronic system that uses infrared beams to measure the limb and calculates the volume based on these measurements. Arm volume is measured preoperatively and at regular intervals corresponding with routine oncology follow-up visits during and after the completion of treatment. This protocol was approved by the Partners HealthCare Institutional Review Board and has been previously published (11).
International Journal of Radiation Oncology Biology Physics Lymphedema was defined as a 10% arm volume increase occurring >3 months postoperatively based on consensus in the literature (12, 13). For patients who underwent unilateral breast surgery, arm volume change was quantified by the relative volume change (RVC) equation (11). Briefly, RVC Z [(A(2)U(1)/U(2) A(1)) 1] where A(1), A(2) are preoperative (1) and postoperative (2) arm volumes on the surgical side and U(1), U(2) are arm volumes on the contralateral side at corresponding time points. This equation accounts for preoperative asymmetry between arms and uses contralateral arm volume as a control to account for factors unrelated to lymphedema, such as weight gain. The RVC equation is not applicable in patients who undergo bilateral breast surgery because a contralateral control arm is lacking. Therefore, the weight-adjusted volume change (WAC) equation was used for patients who underwent bilateral breast surgery (14). Briefly, WAC Z [(A(2)W(1)/W(2)A(1)) 1] where A(1), A(2) are preoperative (1) and postoperative (2) arm volumes on the surgical side, and W(1), W(2) are the patient’s weight at the corresponding time points. The equivalency of 10% according to the RVC and WAC equations as a criterion for lymphedema has been previously demonstrated (14). Arm volume change was calculated at each follow-up visit. Measurements obtained 3 months after surgery were not used to determine lymphedema incidence. A 10% arm volume increase occurring during this period was considered early postoperative swelling and was included as a separate variable in the analysis.
Radiation treatment Of the 1099 breasts that received radiation, 796 (72%) were at the Massachusetts General Hospital (MGH), 92 (8%) at MGHaffiliated institutions, and 211 (19%) at outside institutions. Each breast was categorized as follows: no radiation; PBI, WBI, or chest wall radiation alone (B/CW); or B/CW with RLNR. RLNR was classified as SC only or SC with a PAB. After 2008, supraclavicular level I, II, and III axillary LNs at our institution were contoured most of the time according to the Radiation Therapy Oncology Group atlas (15). The lateral border of the SC field was defined by the treating physician. The SC and level III LNs received a dose of 50 to 50.4 Gy in 25 to 28 fractions (1.82.0 Gy per fraction) through an anterior oblique field angled 10 to 15 to avoid the spinal cord. The dose was calculated at a depth of mostly 3 cm or a depth to include the majority of the contoured supraclavicular area, which varied from 2 to 5 cm. The PAB was completed by use of a posterior approach, and the dose was calculated at midplane. Photon energies of 6- or 10-MV photons were typically used. Until 2009, patients who received PBI were enrolled in a dose-escalation clinical trial (16) and received a total dose of 32, 36, or 40 Gy in 4 Gy per fraction twice daily with a 3-dimensional conformal technique that used a combination of photons and electrons (17). Thereafter, patients were treated off trial with 36 Gy in 9 fractions twice daily. The radiation specifics are given in Table 1.
Statistical analysis Univariate and multivariate analyses were used to identify independent predictors of lymphedema by use of the Cox proportional hazards method. The factors evaluated in these analyses included patient-related factors (age at diagnosis, body mass index [BMI],
Volume - Number - 2014 Table 1
Breast cancer irradiation and lymphedema
Radiation dose and fractionation by treatment field Radiation
Median dose, Gy (range)
Fractions (range)
Partial breast 36.0 (32.0-40.0) 9 Whole breast/chest wall tangents 50.0 (41.9-61.2) 25 Whole breast/chest wall boost 10.0 (2.5-20.0) 5 Supraclavicular 50.0 (42.0-50.4) 25 Posterior axillary boost 46.2 (35.0-48.6) 28
(8-10) (16-28) (3-10) (22-28) (22-28)
and presence of postoperative swelling, defined as a 10% arm volume increase 3 months from surgery); tumor-related factors (tumor size and number of pathologically involved LNs); and treatment-related factors (type of breast surgery, management of the axilla, number of LNs removed, use of systemic therapy, timing of systemic therapy, hormonal therapy, and radiation field [s]). The factors included were based on a review of prior literature regarding lymphedema-associated risk factors. Age at diagnosis, BMI, tumor size, the number of LNs dissected, and the number of pathologically involved LNs were analyzed as continuous variables in such a manner that the hazard ratios computed for these variables reflected the change in lymphedema risk associated with a 1-unit increase in the variable. Timedependent covariates were included for use of systemic therapies and radiation fields in such a manner that cases were included in the unexposed group before the initiation of a given treatment and in the exposed group thereafter. All the aforementioned factors were included in the initial univariate model. Only variables that were significant in the univariate analysis were included in the multivariate analysis. Tests of interaction were performed between the radiation groups and the number of nodes removed, number of positive nodes, and breast surgery, and there were no significant interactions. The Kaplan-Meier method was used to estimate and plot the cumulative incidence of lymphedema by radiation field. Patients were censored at the time of last follow-up measurement.
Results Of the 1476 patients, 1262 (86%) underwent unilateral breast surgery and 214 (14%) underwent bilateral breast surgery for unilateral (nZ189) or bilateral (nZ25) breast cancer. Postoperative radiation therapy was given after 1099 of 1501 (73%) individual breast surgeries. The clinical and pathologic characteristics are given in Table 2. The median postoperative follow-up time was 25.4 months (range, 3.0-82.6 months), with a median of 4 postoperative measurements per patient (range, 1-25 measurements).
Table 2 cohort
3
Clinical and pathologic characteristics of the study
Characteristic
Median (range)
nZ1501 (100%)
Clinical characteristics Age at diagnosis, y 54 (23-89) BMI at diagnosis, kg/m2 26.1 (16.5-59.0) Pathologic characteristics Invasive tumor size,* cm 1.5 (0.01-10.5) Breast surgery Lumpectomy 927 (62) Mastectomy 574 (38) Axillary surgery None 178 (12) SLNB 928 (62) ALND 395 (26) No. of LNs removed, SLNB 2 (1-13) No. of LNs removed, ALND 16 (3-43) No. of positive LNs, ALND 2 (0-39) Radiation therapy None 402 (27) Partial breast 92 (6) B/CW 698 (47) B/CW þ SC 194 (13) B/CW þ SC þ PAB 115 (8) Neoadjuvant chemotherapy Yes 151 (10) No 1350 (90) Adjuvant chemotherapy Yes 571 (38) No 930 (62) Hormonal therapy Yes 1110 (74) No 391 (26) Abbreviations: ALND Z axillary lymph node dissection; B/CW Z whole breast/chest wall only; BMI Z body mass index; LN Z lymph node; PAB Z posterior axillary boost; SC Z supraclavicular; SLNB Z sentinel lymph node biopsy. * Excludes tumor size for patients who underwent neoadjuvant chemotherapy.
The 2-year cumulative incidence of lymphedema by radiation field and axillary surgery is shown in Table 3. For SLNB, the 2year cumulative incidence was 6.1% (95% CI, 1.5%-23.4%) with RLNR (nZ46) and 2.8% (95% CI, 1.6%-4.8%) with PBI/ WBI/chest wall alone (nZ569). For ALND, the incidence was 24.3% (95% CI, 18.9%-30.9%) with RLNR (nZ262) and 7.3% (95% CI, 3.1%-16.8%) with WBI/chest wall alone (nZ73).
Lymphedema incidence Univariate and multivariate analysis The overall cumulative incidence of lymphedema increased from 6.8% (95% confidence interval [CI], 5.5%-8.4%) at 24 months to 13.7% (95% CI, 11.1%-16.9%) at 60 months postoperatively (Fig. 1). The median time to the development of lymphedema was 24.1 months postoperatively (range, 3.0-82.6 months). The 2-year cumulative incidence of lymphedema was higher with RLNR as SC (21.9%; 95% CI, 16.0%-29.5%) or SC þ PAB (21.1%; 95% CI, 13.8%-31.5%) than with no radiation (3.0%; 95% CI, 1.6%-5.6%) or PBI/WBI/chest wall radiation alone (3.1%; 95% CI, 2.0%-4.7%).
The results of univariate analysis are included in Table 4. In a multivariate model using no radiation as the reference group, PBI (PZ.67) and WBI/chest wall radiation alone (PZ.92) did not significantly increase lymphedema risk. Given these results, a multivariate model was constructed with PBI and WBI/chest wall radiation groups combined as the reference group based on perceived clinical relevance. Additionally, because the type of breast surgery was not significant in the multivariate model (PZ.99) and there was not a significant interaction between type
4
International Journal of Radiation Oncology Biology Physics
Warren et al.
Fig. 1. Cumulative incidence of lymphedema stratified by type of radiation therapy. B/CW Z breast/chest wall only; PAB Z posterior axillary boost; SC Z supraclavicular. of breast surgery and radiation, all patients, regardless of surgical type, were included in the final multivariate model. The results of the multivariate analysis are seen in Table 5. The addition of RLNR (SC or SC þ PAB) when compared with breast/chest wall radiation alone significantly increased lymphedema risk, with a hazard ratio (HR) of 1.70 (95% CI, 1.07-2.70, PZ.025). Other factors significant for increased lymphedema risk on multivariate analysis included early postoperative swelling (HR Z 10.3, P<.0001), higher preoperative BMI (HR Z 1.06, P<.0001), number of axillary LNs dissected (HR Z 1.04, PZ.02), and undergoing ALND (HR Z 3.5, PZ.0001).
Discussion The current study used a cohort of 1476 breast cancer patients prospectively screened for lymphedema at a single institution. The patients in our study were treated for breast cancer between 2005 and 2012, and they thereby represent a population treated with modern radiation therapy and surgical techniques over a relatively short time. The risk factors for lymphedema, including adjuvant radiation therapy, have been previously evaluated; however, most studies have used retrospectively collected data, variable and Table 3
largely subjective methods, or both to measure and define lymphedema (12, 18-20). The cumulative incidence of lymphedema in our series was 6.8% at 24 months and 13.7% at 60 months postoperatively. This is lower than has been reported in previous studies, likely because we included a large cohort of women who did not undergo axillary surgery (nZ178). When stratified by radiation therapy field, the 2-year cumulative incidence was 3.0%, 3.1%, 21.9%, and 21.1% for no radiation, breast/chest wall radiation alone, B/CW þ SC, and B/CW þ SC þ PAB, respectively. To evaluate the risk of lymphedema with inclusion of RLNR, we used a reference group composed of patients who received PBI or WBI/chest wall radiation alone, given that RLNR is unlikely to be used without also irradiating the breast or chest wall. In our series, the inclusion of a SC field, with or without the inclusion of a PAB, significantly increased the risk of lymphedema compared with breast/chest wall radiation alone. The addition of a PAB to a SC field did not increase the risk of lymphedema in comparison with a SC field alone. When further stratified by type of axillary surgery, for patients who underwent ALND, the 2-year cumulative incidence was higher with RLNR (as SC or SC þ PAB) than with breast/chest wall radiation alone, 24.3% versus 7.3%, respectively. A similar trend was observed for patients who underwent SLNB (2.8% without RLNR vs 6.1% with RLNR); however, confidence intervals were overlapping because of a small number of patients who underwent SLNB with RLNR (nZ46). These findings are consistent with several prior studies. In a retrospective study of 727 patients treated from 1982 to 1995, Coen et al (19) used a postoperative arm circumference measurement (without preoperative assessment) to evaluate predictors of lymphedema in patients undergoing breast conservation therapy (BCT). The authors found that the addition of RLNR significantly increased the risk of lymphedema from 1.8% to 8.9% with inclusion of RLNR (SC with or without PAB). Another study by Hayes et al (18) evaluated a cohort of 2579 breast cancer patients treated between 1970 and 2005, classifying lymphedema based on physician’s examination, without preoperative assessment. Consistently with our findings, the authors concluded that the inclusion of RLNR significantly increased the risk of lymphedema (PZ.0001), with no significant difference between SC and SC þ PAB (PZ.1). By contrast, Shah et al (20) studied 1497 patients treated from 1980 to 2006 with BCT and did not find that RLNR was a significant risk factor for lymphedema. Increased rates with SC (9.9% vs 7.0%, PZ.29) and PAB (14.7% vs 7.3%, PZ.40) were not observed compared with breast radiation alone. The study did not use a preoperative assessment, and it defined lymphedema based on physical examination with or without circumference measurements. Graham et al (21) evaluated the impact of SC radiation therapy volumes on lymphedema in a cross-sectional study
Cumulative incidence of lymphedema 24 months postoperatively stratified by type of radiation therapy and axillary surgery Cumulative incidence of lymphedema*
Radiation Radiation field(s) No radiation B/CW only B/CW þ SC B/CW þ SC þ PAB
Entire cohort (nZ1501) 3.0% (1.6-5.6) (nZ402) 3.1% (2.0-4.7) (nZ790) 21.9% (16.0-29.5) (nZ194) 21.1% (13.8-31.5) (nZ115)
SLNB 0.00% 2.82% 7.14% 4.76%
(nZ928) (0.00-0.00) (nZ313) (1.7-4.8) (nZ569) (1.0-40.9) (nZ25) (0.7-29.3) (nZ21)
ALND (nZ395) 18.27% (9.7-32.9) (nZ60) 7.33% (3.1-16.8) (nZ73) 23.93% (17.5-32.2) (nZ169) 24.89% (16.2-37.1) (nZ94)
Abbreviations: ALND Z axillary lymph node dissection; B/CW Z partial breast or whole breast/chest wall; PAB Z posterior axillary boost; SC Z supraclavicular; SLNB Z sentinel lymph node biopsy. * 95% confidence interval appears after percentage. No. in each subgroup is shown.
Volume - Number - 2014
Breast cancer irradiation and lymphedema
Table 4 Univariate results for association of factors with risk of lymphedema Variable Clinical characteristics Age at diagnosis* BMI* Pathologic characteristics Invasive tumor size* Breast surgery Mastectomy vs lumpectomy Axillary surgery SLNB vs no axillary surgery ALND vs SLNB No. of LNs removed* No. of positive LNs* Radiation therapyy Partial breast only Whole breast/chest wall only Whole breast/chest wall þ SC Whole breast/chest wall þ SC þ PAB Systemic therapy Neoadjuvant chemotherapy Adjuvant chemotherapy Hormonal therapy Early postoperative swelling 10%, 3 months postoperatively
Hazard 95% Confidence ratio interval
P value
1.01 1.06
0.99-1.02 1.04-1.09
.38 <.0001
1.32
1.22-1.44
<.0001
1.87
1.31-2.67
.0006
0.74
0.31-1.81
.51
9.17 1.10 1.09
5.88-14.29 1.08-1.11 1.05-1.13
<.0001 <.0001 <.0001
0.38 0.85
0.09-1.56 0.51-1.42
.18 .53
5.05
3.11-8.19
<.0001
4.98
2.92-8.47
<.0001
2.61
1.68-4.07
<.0001
2.35 1.68
1.62-3.39 1.13-2.51
<.0001 .01
6.06-26.1
<.0001
12.6
Abbreviations: ALND Z axillary lymph node dissection; BMI Z body mass index; LN Z lymph node; SC Z supraclavicular; SLNB Z sentinel lymph node biopsy; PAB Z posterior axillary boost. * Analyzed as continuous variables; hazard ratio reflects the change in lymphedema risk associated with a 1-unit increase. y Reference group: no radiation therapy.
of 91 breast cancer patients who had postoperative (but no preoperative) arm circumference measurements. They found that SC volumes limited laterally by the coracoid process did not increase the lymphedema risk expected from ALND, whereas expanding the field or including a PAB did. Our cohort uniquely included 402 patients who did not receive radiation, primarily mastectomy patients with limited nodal involvement and older women undergoing lumpectomy. When this cohort was used as a reference group, PBI and WBI/chest wall radiation alone did not significantly increase lymphedema risk. Nonreceipt of radiation therapy is rarely a comparison group in other studies; however, our findings are consistent with those from Liljegren et al (22), which demonstrated that postoperative breast radiation therapy alone did not contribute to lymphedema. It has been suggested that the use of PBI may reduce the rates of lymphedema secondary to lower doses to the axilla (23). Our series demonstrated low rates for patients who underwent BCT with PBI or WBI, whether after SLNB or no axillary surgery. The PBI cohort
5
Table 5 Multivariate results for factors associated with risk of lymphedema Variable No radiation* RLNR* B/CW þ SC B/CW þ SC þ PAB ALND No. of LNs removedy BMIy 10% swelling, 3 months postoperatively
Hazard 95% Confidence ratio interval 0.99 1.70 1.68 1.72 3.52 1.04 1.06 10.3
0.59-1.67 1.07-2.70 1.00-2.83 1.00-2.96 1.84-6.73 1.01-1.07 1.03-1.09 4.52-23.3
P value .9825 .0253 .0492 .0484 .0001 .0184 <.0001 <.0001
Abbreviations: ALND Z axillary lymph node dissection; BMI Z body mass index; LN Z lymph node; RLNR Z regional lymph node radiation. * Radiation reference group: Partial breast and whole breast/chest wall only. y Analyzed as continuous variables; hazard ratio reflects the change in lymphedema risk associated with a 1-unit increase.
in our study was composed of patients who underwent lumpectomy for low-risk, node-negative disease; therefore, further research is warranted to determine whether PBI reduces lymphedema risk compared with WBI for patients with higher-risk or node-positive disease, as in the NSABP B-39/Radiation Therapy Oncology Group 0413 trial (Clinicaltrials.gov ID #: NCT00103181). Other independent risk factors for lymphedema in our series included higher preoperative BMI, ALND, and early postoperative swelling. A previous study by Mahamaneerat et al (24) evaluated postoperative arm swelling as a risk factor for lymphedema and found that a 5% increase in arm volume 1 month after surgery was associated with a 1.4-fold increased risk of lymphedema. Interestingly, in this study, early postoperative swelling conferred the highest risk of lymphedema for all variables analyzed, with a hazard ratio of 10.3. Higher BMI and ALND have also been consistently reported as risk factors in previous studies (8, 12, 25, 26). We found that even after type of axillary surgery was controlled for in a multivariate model, lymphedema risk remained significantly increased with a greater number of LNs removed, which has also been demonstrated in prior studies (18, 25). Although these data are unlikely to change the SLNB or ALND procedure, it may allow for the identification of women for whom closer surveillance is appropriate. The findings from several recent studies may have an impact on practices regarding treatment of the axilla. The ACOSOG Z0011 trial demonstrated no difference in the rates of regional recurrence, DFS, or OS in 891 women with limited nodal involvement receiving SLNB randomized to completion ALND or no further axillary treatment (27, 28). Although it was recommended that patients not receive a third field, they may have had deep or partially deep tangents, which can cover a significant part of the axilla (29). The rates of lymphedema in this trial have not yet been reported. Preliminary findings presented by Rutgers et al (30) on the EORTC 10981-22023 AMAROS [After Mapping of the Axilla: Radiotherapy Or Surgery] phase 3 trial comparing axillary radiation with ALND after a positive SLNB result demonstrated no differences in OS or DFS. The lymphedema rates were higher for ALND (28%) without RLNR than for SLNB with RLNR (14%) at 5 years (P<.0001). Whelan et al (31) reported on the NCIC-CTG MA.20 Intergroup trial, which randomized high-risk women (defined by
6
Warren et al.
positive LN or negative LN with high-risk features) who underwent BCT to receive WBI with or without RLNR. The inclusion of RLNR improved locoregional DFS and distant DFS, and it trended toward a statistically significant improvement in OS. The rate of lymphedema was 4.1% and 7.3% for WBI and WBI þ RLNR, respectively (PZ.004). We similarly observed an increased lymphedema incidence for patients who underwent ALND and RLNR (24.3%) compared with ALND and B/CW radiation alone (7.3%) and/or SLNB and RLNR (6.1%), although these rates include mastectomy patients, unlike the MA.20 trial, and the number of patients in each subgroup is small. A better understanding of the morbidities secondary to combinations of surgical and radiation therapy treatments of the axilla is increasingly important because of the likely change in practice patterns resulting from the publication of these trials, including more frequent use of RLNR. Our study has potential limitations. Potential bias is associated with nonrandom assignment of radiation fields. Additionally, on the basis of practice patterns, the number of patients who underwent SLNB and received RLNR was small, making subgroup analysis challenging and warranting further investigation. Last, the size of the supraclavicular field, and in particular the lateral border, was not examined in the current study. This may have an impact on lymphedema risk and will be the focus of further research. Our study has several strengths. The patients underwent preoperative and regular postoperative measurements by use of a Perometer. The validity and accuracy of the Perometer in quantifying arm volume has been rigorously demonstrated (32). The importance of obtaining preoperative assessments to account for asymmetry between arms, and of incorporating change in contralateral arm size to account for factors unrelated to lymphedema, has also been demonstrated (11, 14, 33, 34). The RVC and WAC equations used in this study account for both of these factors. Previous often-cited studies evaluating radiation therapy-related risk factors for lymphedema have used patients who underwent breast cancer treatment over several decades, and they lack prospectively collected and accurate assessments with incorporation of a preoperative assessment (18-20). By contrast, the current study used patients treated over a 7-year period and reflects modern surgical and radiation therapy techniques, including an increasing use of SLNB as opposed to ALND, and inclusion of patients who received PBI and those with positive SLNB who received RLNR instead of ALND. In conclusion, the addition of RLNR, either with SC or SC þ PAB, significantly increased the risk of lymphedema when compared with breast or chest wall radiation alone. Other independent risk factors included higher preoperative BMI, use of ALND, number of LN dissected, and early postoperative swelling. When considering the inclusion of RLNR, clinicians should weigh the potential benefit of RLNR for control of disease against the increased risk of lymphedema, particularly in women with other potential risk factors for lymphedema. Women who do receive RLNR should be prospectively monitored for lymphedema to ensure early detection and possible intervention. Future studies should further evaluate the risk of lymphedema after SLNB with RLNR compared with completion ALND in the setting of BCT or mastectomy.
References 1. Ahmed RL, Schmitz KH, Prizment AE, et al. Risk factors for lymphedema in breast cancer survivors: The Iowa Women’s Health Study. Breast Cancer Res Treat 2011;130:981-991.
International Journal of Radiation Oncology Biology Physics 2. Ja¨ger G, Do¨ller W, Roth R. Quality-of-life and body image impairments in patients with lymphedema. Lymphology 2006;39:193-200. 3. Tobin MB, Lacey HJ, Meyer L, et al. The psychological morbidity of breast cancer-related arm swelling: Psychological morbidity of lymphoedema. Cancer 1993;72:3248-3252. 4. Khan F, Amatya B, Pallant JF, et al. Factors associated with long-term functional outcomes and psychological sequelae in women after breast cancer. Breast 2012;21:314-320. 5. Ceilley E, Jagsi R, Goldberg S, et al. Radiotherapy for invasive breast cancer in North America and Europe: Results of a survey. Int J Radiat Oncol Biol Phys 2005;61:365-373. 6. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997;337:949-955. 7. Ragaz J, Olivotto IA, Spinelli JJ, et al. Locoregional radiation therapy in patients with high-risk breast cancer receiving adjuvant chemotherapy: 20-year results of the British Columbia randomized trial. J Natl Cancer Inst 2005;97:116-126. 8. McLaughlin SA, Wright MJ, Morris KT, et al. Prevalence of lymphedema in women with breast cancer 5 years after sentinel lymph node biopsy or axillary dissection: Objective measurements. J Clin Oncol 2008;26:5213-5219. 9. Wilke LG, McCall LM, Posther KE, et al. Surgical complications associated with sentinel lymph node biopsy: Results from a prospective international cooperative group trial. Ann Surg Oncol 2006;13:491-500. 10. Langer I, Guller U, Berclaz G, et al. Morbidity of sentinel lymph node biopsy (SLN) alone versus SLN and completion axillary lymph node dissection after breast cancer surgery: A prospective Swiss multicenter study on 659 patients. Ann Surg 2007;245:452-461. 11. Ancukiewicz M, Russell TA, Otoole J, et al. Standardized method for quantification of developing lymphedema in patients treated for breast cancer. Int J Radiat Oncol Biol Phys 2011;79:1436-1443. 12. Tsai RJ, Dennis LK, Lynch CF, et al. The risk of developing arm lymphedema among breast cancer survivors: A meta-analysis of treatment factors. Ann Surg Oncol 2009;16:1959-1972. 13. Armer JM, Stewart BR. A comparison of four diagnostic criteria for lymphedema in a post-breast cancer population. Lymphat Res Biol 2005;3:208-217. 14. Miller CL, Specht MC, Horick N, et al. A novel, validated method to quantify breast-cancer related lymphedema (BCRL) following bilateral breast surgery. Lymphology 2013;46:64-74. 15. Li XA, Tai A, Arthur DW, et al. Variability of target and normal structure delineation for breast cancer radiotherapy: An RTOG multiinstitutional and multiobserver study. Int J Radiat Oncol Biol Phys 2009;73:944-951. 16. Pashtan IM, Recht A, Ancukiewicz M, et al. External beam accelerated partial-breast irradiation using 32 gy in 8 twice-daily fractions: 5year results of a prospective study. Int J Radiat Oncol Biol Phys 2012; 84:e271-e277. 17. Taghian AG, Kozak KR, Katz A, et al. Accelerated partial breast irradiation using proton beams: Initial dosimetric experience. Int J Radiat Oncol Biol Phys 2006;65:1404-1410. 18. Hayes SB, Freedman GM, Li T, et al. Does axillary boost increase lymphedema compared with supraclavicular radiation alone after breast conservation? Int J Radiat Oncol Biol Phys 2008;72:1449-1455. 19. Coen JJ, Taghian AG, Kachnic LA, et al. Risk of lymphedema after regional nodal irradiation with breast conservation therapy. Int J Radiat Oncol Biol Phys 2003;55:1209-1215. 20. Shah C, Wilkinson JB, Baschnagel A, et al. Factors associated with the development of breast cancer-related lymphedema after whole-breast irradiation. Int J Radiat Oncol Biol Phys 2012;83:1095-1100. 21. Graham P, Jagavkar R, Browne L, et al. Supraclavicular radiotherapy must be limited laterally by the coracoid to avoid significant adjuvant breast nodal radiotherapy lymphoedema risk. Australas Radiol 2006; 50:578-582. 22. Liljegren G, Holmberg L. Arm morbidity after sector resection and axillary dissection with or without postoperative radiotherapy in breast
Volume - Number - 2014
23. 24.
25.
26.
27.
28.
cancer stage I. Results from a randomised trial. Uppsala-Orebro Breast Cancer Study Group. Eur J Cancer 1997;33:193-199. Shah C, Badiyan S, Khwaja S, et al. Breast cancer related lymphedema: A review of recent developments. Androl Gynecol Curr Res 2013;1:2. Mahamaneerat WK, Shyu C-R, Stewart BR, et al. Breast cancer treatment, BMI, post-op swelling/lymphoedema. J Lymphoedema 2008;3:38-44. Disipio T, Rye S, Newman B, et al. Incidence of unilateral arm lymphoedema after breast cancer: A systematic review and metaanalysis. Lancet Oncol 2013;14:500-515. Lucci A, McCall LM, Beitsch PD, et al. Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J Clin Oncol 2007; 25:3657-3663. Giuliano AE, Hunt KK, Ballman KV, et al. Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: A randomized clinical trial. JAMA 2011;305:569-575. Giuliano AE, McCall L, Beitsch P, et al. Locoregional recurrence after sentinel lymph node dissection with or without axillary dissection in patients with sentinel lymph node metastases: The American College of Surgeons Oncology Group Z0011 randomized trial. Ann Surg 2010; 252:426-432. discussion 432-433.
Breast cancer irradiation and lymphedema
7
29. Haffty BG, Hunt KK, Harris JR, et al. Positive sentinel nodes without axillary dissection: Implications for the radiation oncologist. J Clin Oncol 2011;29:4479-4481. 30. Rutgers EJ, Donker M, Straver ME, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer patients: Final analysis of the EORTC AMAROS trial (10981/22023) [Abstract]. In: Proceedings of the ASCO Annual Meeting; 2013 May 31-June 4; Chicago, IL. J Clin Oncol 2013; 31 (Suppl.):Abstr LBA1001. 31. Whelan T, Olivotto I, Chapman J. NCIC-CTG MA.20: An intergroup trial of regional nodal irradiation in early breast cancer [Abstract]. In: Proceedings of the ASCO Annual Meeting; 2011 June 3-June 7; Chicago, IL. J Clin Oncol 2011;29 (Suppl.):Abstr LBA 1003. 32. Tierney S, Aslam M, Rennie K, et al. Infrared optoelectronic volumetry: The ideal way to measure limb volume. Eur J Vasc Endovasc Surg 1996;12:412-417. 33. Stout Gergich NL, Pfalzer LA, McGarvey C, et al. Preoperative assessment enables the early diagnosis and successful treatment of lymphedema. Cancer 2008;112:2809-2819. 34. Ancukiewicz M, Miller CL, Skolny MN, et al. Comparison of relative versus absolute arm size change as criteria for quantifying breast cancer-related lymphedema: The flaws in current studies and need for universal methodology. Breast Cancer Res Treat 2012;135:145-152. erratum in Breast Cancer Res Treat 2012;136:623.